[2102.08790] Theory and Functionals in the Hypercomplex Kohn-Sham DFT Method

The static/strong correlation error inherent in existing functionals has severely limited the further application of Kohn-Sham (KS) density functional theory (DFT). The recently developed hypercomplex KS (HCKS) theory shows great potential to overcome the fundamental limitations of the conventional KS-DFT, hence further development and application of HCKS will effectively guide the construction of new functionals toward better description of strong correlation. To this end, this work derives the working equations for the HCKS calculation and proves that HCKS using common functionals has much lower computational complexity than reduced density matrix functional theory (RDMFT), while maintaining the same density search space as RDMFT. Moreover, this work combines HCKS with hybrid functionals that are based on orbitals and their occupations. The test on triplet-singlet gaps shows that the systematic error of HCKS due to the lack of static correlation in approximate functionals can be effe

1 mentions: @KwhRd100
Date: 2021/02/20 15:51

Referring Tweets

@KwhRd100 ハイパーコンプレックスコーン-シャムDFT法における理論と汎関数 HCKSの計算複雑度は、RDMFTで用いている軌道エネルギーの最小化勾配アルゴリズムを適用する代わりに、固有値方程式を解くことで大幅に低減できる。HCKSの系統誤差は、静的相関補正を含めることで低減できる。t.co/8w2nCPmeuU

Related Entries

Read more Kaggle で10年遊んだ GrandMaster の振り返り
183 users, 26 mentions 2021/02/18 02:21
Read more BERTモデルを用いた日本語係り受け解析ツールSuPar-UniDicリリース | yasuokaの日記 | スラド
2 users, 2 mentions 2021/02/22 09:51
Read more 経済産業省のAIコンペに参加しました|Norikazu|note
0 users, 2 mentions 2021/02/22 23:21
Read more NFNet(バッチ正規化を用いない高精度画像認識)の元論文を解説 | 医師によるAI開発
1 users, 2 mentions 2021/02/21 00:46
Read more GitHub - vaaaaanquish/Awesome-Rust-MachineLearning: Awesome Rust Machine Learning crate list
0 users, 0 mentions 2021/02/25 12:53